Strategies toward improving the brain penetration of macrocyclic tertiary carbinamine BACE-1 inhibitors

Article abstract of DOI:10.1016/j.bmcl.2007.08.040

This letter describes replacements for the P3 amide moiety present in previously reported tertiary carbinamine macrolactones. Although P-gp efflux issues associated with these amide-macrolactones were solved and full brain penetration was measured in one

Full text of DOI:10.1016/j.bmcl.2007.08.040

Bioorganic & Medicinal Chemistry Letters 17 (2007) 5831–5835
Strategies toward improving the brain penetration of macrocyclic
tertiary carbinamine BACE-1 inhibitors
Keith P. Moore,a,* Hong Zhu,a Hemaka A. Rajapakse,a Georgia B. McGaughey,b
Dennis Colussi,c Eric A. Price,c Sethu Sankaranarayanan,c Adam J. Simon,c
Nicole T. Pudvah,d Jerome H. Hochman,d Timothy Allison,e Sanjeev K. Munshi,e
Samuel L. Graham,a Joseph P. Vaccaa and Philippe G. Nantermeta
aDepartment of Medicinal Chemistry, Merck Research Laboratories, PO Box 4, West Point, PA 19486, USA
bDepartment of Molecular Systems, Merck Research Laboratories, PO Box 4, West Point, PA 19486, USA
cDepartment of Alzheimer’s Research, Merck Research Laboratories, PO Box 4, West Point, PA 19486, USA
dDepartment of Drug Metabolism, Merck Research Laboratories, PO Box 4, West Point, PA 19486, USA
eDepartment of Structural Biology, Merck Research Laboratories, PO Box 4, West Point, PA 19486, USA
Received 3 July 2007; revised 17 August 2007; accepted 17 August 2007
Available online 23 August 2007
Abstract—This letter describes replacements for the P3 amide moiety present in previously reported tertiary carbinamine macrolac-
tones. Although P-gp efflux issues associated with these amide-macrolactones were solved and full brain penetration was measured
in one case, potency was compromised in the process.
ꢀ 2007 Elsevier Ltd. All rights reserved.
Alzheimer’s Disease (AD) is a neurodegenerative disease
of the brain that leads to a progressive decline in cogni-
tive function and ultimately incapacitation and death.1
AD is characterized by the presence of insoluble amy-
loid plaques and fibrillary tangles, which are key patho-
logical features of this devastating disease. The major
component of these plaques is Ab40À42, a neurotoxic
peptide fragment of b-amyloid precursor protein
(APP) generated by the proteolytic action of the b-
and c-secretase enzymes. Of these two enzymes, b-secre-
tase (b-site APP Cleaving Enzyme or BACE-1), an
aspartyl protease, is considered rate-limiting in this pro-
teolytic cleavage process.2 As such, BACE-1 inhibition
is widely hypothesized to be one of the most promising
therapeutic approaches for the treatment of AD.3
state isosteres such as hydroxyethylamines (HEA), re-
duced amides, statines, and aminostatines to interact
with the catalytic aspartates.3 The incorporation of these
motifs into high molecular weight inhibitors negatively
impacts pharmacokinetics and penetration of the blood
brain barrier (BBB). The discovery of non-transition
state isostere tertiary carbinamine derived BACE-1 li-
gands was recently disclosed.4 Inhibitors based on this
novel ‘warhead’ motif displayed comparable potency
P2
Ms
R
N
Z
P3
O
Y
X
O
O
Ph
Gly-230
NH
The development of low molecular weight, brain pene-
trant BACE-1 inhibitors has been extremely challeng-
ing.3 Most efforts have relied on the use of transition
O
O
R
NH2
NH2
1a: R = MeNMs:
BACE-1 IC50 = 2 nM
2-4
Keywords: Alzheimer’s disease; BACE-1; Glycine 230; Tertiary
carbinamine; P-gp efflux pump; Macrolactone; 10s loop; Macrocyclic
N-arylation; Negishi coupling.
1b: R = 2-CN-Ph:
BACE-1 IC50 = 27 nM
*
Figure 1. Macrolactone BACE inhibitor 1a and proposed P3 amide
replacements.
Corresponding author. Tel.: +1 215 652 6976; e-mail:
keith_moore2@merck.com
0960-894X/$ - see front matter ꢀ 2007 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmcl.2007.08.040

5832
K. P. Moore et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5831–5835
Ms
Ms
to classical HEA inhibitors although stability, pharma-
cokinetics and brain penetration remained an issue.
Macrolactones such as 1a,b (Fig. 1) derived from this
new scaffold displayed a large increase in potency and
stability but no significant improvement in pharmacoki-
netics and BBB penetration.5 Additional work in this
series indicated that the P3 amide moiety, necessary
for making a critical interaction with BACE-1 Gly230,
could be altered to afford molecules with similar potency
and reduced P-gp liabilities.6,11 Herein is described our
efforts to replace the P3 amide in an attempt to improve
BBB penetration.
N
N
N
F
Y
N
Y
O
NH
O
O
NH
NH2
NH2
R
hu P-gp
hu P-gp
BACE-1
Y
BACE-1
Y
R
B-A/A-B6
B-A/A-B6
IC50 (nM)7
IC50 (nM)7
H
F
CO
CO
1c
1a
2
9
16
>50
3.2
1d CHCF3
2a CHCF3 3700
15000
1.3
Figure 2. CF3 replacement of carbonyl in both acyclic and macrocyclic
modes.
Our first strategy relied on the replacement of the P3
amide with a trifluoroethylamine moiety, following
precedent reported by Blacket al.8 Previous crystallo-
graphic analyses demonstrated that the P3 carbonyl does
not interact directly with the enzyme,4 which should al-
low for its replacement without consequences on inhib-
itory potency. The interaction of the amide NH with
Gly230 could be maintained via an amide surrogate.
The trifluoromethyl group has been shown to have an
electronic effect similar to that of the carbonyl on the
pKa of the NH,8 and should therefore not affect its crit-
ical H-bonding potential to Gly230. Furthermore, the
amide carbonyl represents a classical recognition ele-
ment to the P-gp efflux pump, its replacement with a tri-
fluoromethyl group might decrease the compound’s
susceptibility to P-gp. Figure 2 illustrates the replace-
ment of the P3 amide carbonyl with a trifluoromethyl
group and its transport by P-gp (2a: B À A/
A À B = 1.3 vs. 1a: B À A/A À B = 16);9 unfortunately
this is achieved at the cost of potency. The accompany-
ing sp2–sp2 to sp3–sp3 hybridization switch in the con-
displayed a significant reduction in P-gp susceptibil-
ity,9,10 it was also considerably less potent than either
carbonyl containing compounds 1a or 1c.
Replacement of the P3–P2 carboxamide-phenyl combi-
nation with a P3–P2 cyclopropylamino-pyridine scaffold
has been shown to reduce P-gp susceptibility while
maintaining adequate potency11and was considered as
a second strategy. Indeed aminopyridine 3a maintains
modest potency at 400 nM with a concomitant lowering
of P-gp susceptibility (B À A/A À B = 2.9, Table 1).
N-Alkylation was then considered to simultaneously
occupy S3 and remove the NH that could serve as a
recognition element for the P-gp efflux pump. However,
N-substituted
aminopyridine-macrolactones
3b–d
suffered a significant loss in potency, indicating that S3
access directly from the a-position is suboptimal.
A third strategy involved the removal of any heteroatom
from the P2–P3 linker and accessing S3 from the more
traditional c-position.5 The concept was initially applied
in the P2 phenyl mode with a P3 methyl, providing mac-
text of
a
strained macrocyclic ring may have
unfavorably altered the vector of the P3 NH toward
Gly230. In order to allow for more conformational flex-
ibility, the acyclic analog 1d was evaluated. Although 1d
rolactone 4a, as
a
non P-gp substrate (B À A/
Table 1. Macrolactone optimization
Ms
N
Z
α
Y
X
O
O
R
γ
NH2
hu P-gpd
a
BACE-1 IC50 (nM)
Compound
R
–X–Y–
Z
PApp
Ratio
3a
3b
3c
3d
4a
4b
4c
4d
H
–CH2–NH–
N
N
N
N
C
400b
13000b
22000b
6000b
1700c
11
20
ND
8
2.9
5.3
ND
0.8
1.4
0.4
0.4
0.8
H
–CH2–NBn–
–CH2–NiPr–
–CH2–NCH2cyPr–
–CH2–CH2–
–CH2–CH2–
–CH2–CH2–
–CH2–CH2–
H
H
Me
Me
Ph
H
16
14
9
N
N
N
1800
7000c
15000b
29
a IC50 is an average of at least three measurements.7
b Calculated from a mixture of 2 enantiomers.12
c Calculated from a mixture of 4 diastereoisomers.12
d PApp in 10À6 cm/s; Efflux ratio: B À A/A À B.6

K. P. Moore et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5831–5835
5833
X-ray crystallography was used to compare the binding
of ligands 4b and 1b and to help rationalize these results.
In the absence of heteroatoms in the P2–P3 linker of li-
gands 4b–d, there can be no hydrogen bonding interac-
tion with BACE-1 Thr232, giving Thr232 the
opportunity to hydrogen bond to Ser10 and thus induc-
ing a 10s loop ‘‘down’’ configuration (Fig. 3a). As pre-
dicted15 and as observed earlier,11 a small hydrophobic
P3 group such as the methyl exemplified in 4a,b is the
best fit for the small S3 pocket in this 10s loop ‘‘down’’
configuration. The larger P3 phenyl (4d) likely maintains
the 10s loop in an ‘‘up’’ configuration that is non-opti-
mal in such cases where the ligand cannot H-bond with
Thr232. In the case of P3 amide inhibitors such as 1b,5
Thr232 is engaged by the amide carbonyl resulting in a
10s loop ‘‘up’’ configuration and a larger S3 pocket
which is best occupied by a P3 phenyl (Fig. 3b).5,15
Figure 3a. X-ray of inhibitor 4b complexed with BACE-1. Flap was
removed for clarity. PDB ID code: 2PH8.
Ms
Ms
Ms
N
N
N
d
A À B = 1.4) displaying moderate potency against
BACE-1 (IC50 = 1700 nM). Encouraged by this result,
we applied this strategy to the P2 pyridyl analog and re-
solved all four diastereoisomers. Enantiomer 4b was
evaluated as a 1800 nM inhibitor. X-ray crystallo-
graphic analysis of 4b bound to BACE-1 (Fig. 3a) con-
firmed4,5 the R configuration at the tertiary carbinamine
center; however, the absolute configuration at the c-po-
sition was unexpectedly found to be S.13 Gratifyingly, 4b
was found to effectively penetrate the BBB (at 2 h,
[4b]brain = 637 nM, [brain]/[plasma] = 130%).14 Guided
by earlier SAR in the P3 amide mode,5 the previously
optimal P3 phenyl was incorporated (4c) resulting in a
surprising decrease in potency. Complete removal of
the P3 group (4d) resulted in further loss of potency.
a-c
F3C
F3C
Ph
CO2t-Bu
CO2t-Bu
CO2t-Bu
OTf
NH
OH
7
5
6
i-k
e-h
Ms
Ms
Ms
N
N
N
NHBoc
NH
N
F3C
F3C
Ar
F3C
Ph
l-o
N
O
NH
O
O
Ar
NH
NH
O
NH2
NH2
Ph
1d
2a
8
Scheme 1. Reagents: (a) TPAP, NMO, 74%; (b) CF3–TMS, cat.
TBAF, 87%; (c) Tf2O, lutidine, 81%; (d) K2CO3, (R)-1-phenylprop-2-
en-1-amine, 84%; (e) 9-BBN, NaOH, (Ph3P)4Pd, 9, 71%; (f) 1—TFA,
DCM, 99%; 2—Boc2O, hunig’s base, 92%; 3—BH3–THF, 69%;
(g) LiOH, MeOH, 99%; (h) 1—DIAD, PPh3, THF, 85%; 2—TFA,
DCM, 99%; (i) K2CO3, (R)-1-(4-fluorophenyl)ethanamine, 88%; (j)
TFA, DCM, 99%; (k) Boc-hydrazine, EDC, HOAt, 66%; (l) TFA,
DCM, 99%; (m) Boc-a-methyl-D-Phe, EDC, HOAt, 73%; (n) CBr4,
PPh3, DCM, 88%; (o) TFA, DCM, 99%.
OH
N3
a,b
c,d
CO2Me
NHBoc
CO2Me
NHBoc
CO2Me
NHBoc
Br
10
9
11
e-g
Ms
Ms
Ms
N
N
N
N
N
N
R
R
N
N
NH2 Cl
j
h,i
or i
O
O
O
O
O
O
NH2
NHBoc
NHBoc
3a-d
13
12
Scheme 2. Reagents: (a) 9-BBN, NaOH, (Ph3P)4Pd, but-3-en-1-yloxy-
t-butyldimethyl silane, 68%; (b) TBAF, THF, 99%; (c) CBr4, PPh3,
imid., 85%; (d) NaN3, DMF, 83%; (e) LiOH, MeOH, 99%; (f) Cs2CO3,
22, DMF, 91%; (g) PPh3, H2O, THF, 70%; (h) for 3b-d: RCHO,
NaBH(OAc)3, 60–80%; (i) for 3a–d: [(t-Bu)3P]2Pd, K3PO4, DMA, 65–
82%; (j) TFA, DCM, 99%.
Figure 3b. X-ray of inhibitor 1b5 complexed with BACE-1. Flap was
removed for clarity. PDB ID code: 2QZK.

5834
K. P. Moore et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5831–5835
Ms
structed from the hydroboration of alkene 14 followed
N
by Suzuki-Miyaura coupling to halides 20 or 21. After
silyl deprotection and ester hydrolysis, macrocyclization
was executed using an intramolecular Mitsunobu reac-
tion which provided macrolactones 4a,b,d in excellent
yields.
Z
OTBS
R
a
CO2Me
NHBoc
b or c
R
CO2Me
NHBoc
R
13
14
15
d-f
For compound 4c, bis-carboxylic acid 16 was used in
place of the corresponding diene to avoid double bond
migration that would generate styrenes. Mono-esterifi-
cation of 16 followed by borane reduction, and conver-
sion to the alkyl iodide provided the corresponding
ester-iodide. The resulting iodide was activated using
Rieke zinc and utilized in a Negishi coupling with pyr-
idyl chloride 21 to give ester 18. Reduction of the ester,
conversion to the alkyl iodide and Rieke zinc activation
enabled a second Negishi coupling with aryl bromide 9
to yield 19, which underwent macrocyclization using
the established protocol to afford macrolactone 4c.
Ms
N
Z
HO2C
CO2H
g-i
CO2Me
I
R
O
O
R
R
16
17
NH2
4a-e
j
d-f
Ms
Ms
Ms
N
N
N
Z
Z
N
k-m
OTBS
Y
OTBS
R
R
OTBS
CO2Me
NHBoc
In conclusion, various replacements for the P3 amide
functionality embedded in tertiary carbinamine macro-
lactones were discussed. Although P-gp efflux issues
were solved in most cases and brain penetration was at-
tained in vivo for 4b, potency was compromised in the
process. Optimization of S3 access from the aminopyri-
dine scaffold 3a will be the subject of future work, to be
described in due time.
20: Y = Br, Z = C
21: Y = Cl, Z = N
CO2Me
18
19
Scheme 3. Reagents: (a) 9-BBN, NaOH, (Ph3P)4Pd, 9, 72%; (b) for 4a:
9-BBN, NaOH, (Ph3P)4Pd, 20, 78%; (c) for 4b, 4d: 9-BBN, NaOH,
(Ph3P)4Pd, 21, 73%; (d) TBAF, THF, 99%; (e) LiOH, MeOH, 99%;
(f) DIAD, PPh3, THF, 88%; (g) MeI, Cs2CO3, DMF, 55%; (h) BH3–
THF, 69%; (i) I2, PPh3, imid., 77%; (j) Reike Zinc, [(t-Bu)3P]2Pd, 21,
82%; (k) LiBH4, THF, 86%; (l) I2, PPh3, imid., 91%; (m) Reike Zinc,
[(t-Bu)3P]2Pd, 9, 78%.
Acknowledgments
The authors thank the Laboratory Animal Resource
and Drug Metabolism departments for animal dosing
and sample analysis; the analytical chemistry, mass spec-
trometry, and NMR groups for analytical and spectro-
scopic data; Carl Homnick for chiral separations; and
Drs. J. Barrow, I. Hills, and W. Trotter for proofreading
of this manuscript.
The synthetic routes16 toward trifluoromethyl deriva-
tives 2a and 1d are outlined in Scheme 1. Benzyl alcohol
5 was oxidized to the corresponding aldehyde, which
then underwent nucleophilic addition of CF3–TMS in
the presence of catalytic TBAF.17 The resulting alcohol
was then converted to triflate 6, and amine displacement
afforded 7. Hydroboration of 7 followed by Suzuki-
Miyaura coupling to aryl bromide 9 (from Scheme 2)
sets the key molecular framework. The synthesis of
2aculminated in a macrolactonization under Mitsunobu
conditions. For 1d, (R)-1-(4-fluorophenyl)ethanamine
was employed to displace triflate 6; formation of the
oxadiazole ring was accomplished under dehydrative
conditions using carbon tetrabromide and triphenyl-
phosphine. Boc deprotection afforded inhibitor 1d.
Supplementary data
Experimental procedures for the synthesis of com-
pounds 2–4. Supplementary data associated with this
article can be found, in the online version, at
doi:10.1016/j.bmcl.2007.08.040.
The syntheses16 of inhibitors 3a–d are outlined in
Scheme 2. Hydrolysis of methyl ester 11 was followed
by carboxylate alkylation with 22, and reduction of
the azide moiety with triphenylphosphine afforded
amine 12. With the ester bond intact, macrocyclization
could be executed using a palladium mediated macrocy-
clic N-arylation18 was followed by protecting group re-
moval to give inhibitor 3a. The macrocyclization could
also be achieved using secondary amines, resulting from
reductive aminations with the necessary aldehydes to
give inhibitors 3b–d.
References and notes
1. Nussbaum, R. L.; Ellis, C. E. N. Eng. J. Med. 2003, 348,
1356; Stachel, S. J.; Vacca, J. P. Curr. Topic Med. Chem.
2006, 6, 551, and following articles.
2. Cummings, J. L. N. Eng. J. Med. 2004, 351, 56.
3. Roberds, S. L.; Anderson, J.; Basi, G.; Bienkowski, M. J.;
Branstetter, D. G.; Chen, K. S.; Freedman, S.; Frigon, N.
L.; Games, D.; Hu, K.; Johnson-Wood, K.; Kappenman,
K. E.; Kawabe, T.; Kola, I.; Kuehn, R.; Lee, M.; Liu, W.;
Motter, R.; Nichols, N. F.; Power, M.; Robertson, D. W.;
Schenk, D.; Schoor, M.; Shopp, G. M.; Shuck, M. E.;
Sinha, S.; Svensson, K. A.; Tatsuno, G.; Tintrup, H.;
Wijsman, J.; Wright, S.; McConlogue, L. Hum. Mol.
Genet. 2001, 10, 1317; Durham, T.; Shepherd, T. Curr.
Macrolactones 4a,b,d were prepared16 as outlined in
Scheme 3. A fully saturated carbon linker could be con-

K. P. Moore et al. / Bioorg. Med. Chem. Lett. 17 (2007) 5831–5835
5835
Opin. Drug Dis. Dev. 2006, 9, 776; Guo, T.; Hobbs, D. W.
Curr. Med. Chem. 2006, 13, 1811; Ziora, Z.; Kimura, T.;
Kiso, Y. Drug Future 2006, 31, 53; Beher, D.; Graham, S.
L. Expert Opin. Investig. Drugs 2005, 14, 1385; Gosh, A.
K.; Kumaragurubaran, N.; Tang, J. Curr. Topics Med.
Chem. 2005, 5, 1609; Thompson, L. A.; Bronson, J. J.;
Zusi, C. Curr. Pharma. Des. 2005, 11, 3383.
11. Stauffer, S. R.; Stanton, M. G.; Gregro, A. R.; Steinbeiser,
M. A.; Nantermet, P. G.; Barrow, J. C.; Rittle, K. E.;
Colussi, D.; Espeseth, A. S.; Lai, M. T.; Pietrak, B.;
Holloway, M. K.; McGaughey, G. B.; Munshi, S.;
Hochman, J. H.; Simon, A. J.; Selnick, H. G.; Graham,
S.; Vacca, J. P. Bioorg. Med. Chem. Lett. 2007, 17, 1788.
12. The stereochemistry of the active diastereoisomer is
known to always be R at the tertiary carbinamine center,
based on earlier X-ray crystallographic analysis of tertiary
carbinamine/P3-amide based acyclic and macrocyclic
inhibitors Refs. 4,5. Additionally, chiral separations per-
formed earlier on similar mixture of diastereoisomers
indicated that the diastereoisomers derived from the S
tertiary carbinamine are all devoid of significant activity
against BACE-1, thus justifying IC50 calculations from
mixtures Refs. 5,12.
13. The R configuration at the c-position for inhibitor 4b is
opposite to what was previously observed with P3 amide
tertiary carbinamine inhibitors (S) Ref. 5. This inversion
of configuration is likely allowed by the increased confor-
mational flexibility introduced in the chain when switching
from an amide to ethylene linker. Only one of the other
three diastereoisomers displayed inhibitory potency at
17 lM. This isomer most likely has the correct R config-
uration at the tertiary carbinamine center and the S
configuration at the c-position.
4. Rajapakse, H. R.; Nantermet, P. G.; Selnick, H. G.;
Munshi, S.; McGaughey, G. B.; Lindlsey, S. R.; Young,
M. B.; Lai, M. T.; Espeseth, A. S.; Shi, X. P.; Colussi, D.;
Pietrak, B.; Crouthamel, M. C.; Tugusheva, K.; Huang,
Q.; Xu, M.; Simon, A. J.; Kuo, L.; Hazuda, D. J.;
Graham, S.; Vacca, J. P. J. Med. Chem. 2006, 49, 7270.
5. Lindsley, S. R.; Moore, K. P.; Rajapakse, H. R.; Selnick,
H. G.; Young, M. B.; Zhu, H.; Munshi, S.; Kuo, L.;
McGaughey, G. B.; Colussi, D.; Crouthamel, M. C.; Lai,
M. T.; Pietrak, B.; Price, E. A.; Sankaranarayanan, S.;
Simon, A. J.; Seabrook, G. R.; Hazuda, D. J.; Pudvah, N.
T.; Hochman, J. H.; Graham, S.; Vacca, J. P.; Nantermet,
P. G. Bioorg. Med. Chem. Lett. 2007, 17, 4057.
6. Hochman, J. H.; Yamazaki, M.; Ohe, T.; Lin, J. H. Curr.
Drug Met. 2002, 3, 257; Yamazaki, M.; Neway, W. E.;
Ohe, T.; Chen, I-W.; Rowe, J. F.; Hochman, J. H.; Chiba,
M.; Lin, J. H. J. Pharmacol. Exp. Ther. 2001, 296, 723.
7. Shi, X.-P.; Tugusheva, K.; Bruce, J. E.; Lucka, A.; Chen-
Dodson, E.; Hu, B.; Wu, G.-X.; Price, E.; Register, R. B.;
Lineberger, J.; Miller, R.; Tang, M.-J.; Espeseth, A.;
Kahana, J.; Wolfe, A.; Crouthamel, M.-C.; Sankarana-
rayanan, S.; Simon, A.; Chen, L.; Lai, M.-T.; Pietrak, B.;
DiMuzio, J.; Li, Y.; Xu, M.; Huang, Q.; Garsky, V.;
Sardana, M. K.; Hazuda, D. J. J. Alzheimer’s Dis. 2005, 7,
139.
14. Brain penetration was assessed from mice brain homog-
enates, two hours post i.p. dosing in 25% DMSO/75%
PEG300, at 30 mpk.
15. McGaughey, G. B.; Colussi, D.; Graham, S. L.; Lai, M.;
Munshi, S.; Nantermet, P. G.; Pietrak, B.; Rajapakse, H.
A.; Selnick, H. G.; Stauffer, S. R.; Holloway, M. K.
Bioorg. Med. Chem. Lett. 2007, 17, 1117.
8. Black, W. C.; Bayly, C. I.; Davis, D. E.; Desmarais, S.;
´
´
Falgueyret, J.; Leger, S.; Li, C.; Masse, F.; McKay, D. J.;
Palmer, J. T.; Percival, M. D.; Robichaud, J.; Tsou, N.;
Zamboni, R. Bioorg. Med. Chem. Lett. 2005, 15, 4741.
9. Compounds 2a and 1d were evaluated as 1:1 diastereo-
meric mixtures. PApp > 20 · 10À6 cm/s for both Ref. 6.
10. The 4-fluoro substitution on the P1 aryl group is known to
improve inhibitory potency against BACE-1 by 2-fold at
best (unpublished results) and is expected to have no
significant impact on P-gp susceptibility.
16. For experimental details, see Supplemental material,
Nantermet, P. G.; Rajapakse, H. A.; Selnick, H. G.;
Lindsley, S. R.; Moore, K. P.; Stachel, S. J.
WO2006055434, and Rajapakse, H. A.; Nantermet, P.
G.; Selnick, H. G.; Moore, K. P.; WO2006057983.
17. Nelson, D. W.; Owens, J.; Hiraldo, D. J. Org. Chem. 2001,
66, 2572.
18. Stauffer, S. R.; Steinbeiser, M. A. Tetrahedron Lett. 2005,
46, 2571.